Anna Nordén Essays on Behavioral Economics and Policies for Provision of Ecosystem Services ________________________ ECONOMIC STUDIES DEPARTMENT OF ECONOMICS SCHOOL OF BUSINESS, ECONOMICS AND LAW UNIVERSITY OF GOTHENBURG 211

(1)

ECONOMIC STUDIES

DEPARTMENT OF ECONOMICS

SCHOOL OF BUSINESS, ECONOMICS AND LAW

UNIVERSITY OF GOTHENBURG

211 ________________________

Essays on Behavioral Economics and Policies for Provision of Ecosystem Services

(2)

(3)

(4)

(5)

Acknowledgements

One of my supervisors once told me that finishing a PhD requires 60% hard work and devotion, 30% social networking, and 10% being smart. Even though this distribution can be discussed, all parts have definitely been involved in the production of this PhD thesis. Research is hard work, and without devotion, who would spend weekends and nights preparing folders for experiments or running LimDep? Further, I am very sure I would never even have started the PhD route without the incredible people that I am blessed with in my life. This yellow book is as much the end as the beginning of something, but before my journey continues I would like to express my gratitude to all of those who have supported me in the completion of this thesis.

First, I want to express my genuine gratitude and admiration to my supervisors: Francisco Alpízar and Peter Martinsson. Francisco and Peter are brilliant researchers and incredible persons with a unique sense for interesting research. Their wisdom, encouragement, and insightful comments have guided me through the long process of writing this thesis. In addition to a lot of interesting research, both Peter and Francisco have always been excellent mentors. It has been a great privilege to work with both of them and become their co-author. Tusen tack Peter! Mil gracias Fran!

My sincere gratitude is also extended to my co-authors Alexander Pfaff, Juan Robalino, and Martin Persson. Their expertise, enthusiasm, and insight have been a great inspiration in my research. I would also like to thank my colleagues at the Department of Economics at the University of Gothenburg for sharing their time with me by being my teachers. Fredrik Carlsson, Olof Johansson-Stenman, Thomas Sterner, Gunnar Köhlin, Håkan Eggert, Åsa Löfgren, Jessica Coria, Måns Söderbom, Arne Bigsten, Ola Olsson, Katarina Nordblom, Johan Stennek, Matthias Sutter, Martin Kocher, Bo Sandelin, Hans Bjurek, Evert Köstner, Lennart Flood, and Renato Aguilar have all taught me a great deal about research and various economic tools. In addition, I greatly appreciate the hospitality and support from the people at the Beijer International Institute of Ecological Economics at the Swedish Academy of Science during the specialization course.

(8)

Environment for Development Initiative (EfD)-Central America center at CATIE in Costa Rica, and Jeanette for scanning and sending my grades several times. I also wish to thank Debbie Axlid for excellent editorial assistance.

Speaking of the EfD-Central America center in Costa Rica, I am fortunate to have had the opportunity to spend more than three years with them. They have all been a great support not only for this thesis but also during the work with my Master’s thesis. The first time I came to CATIE in 2007, the environmental economics unit consisted of Francisco Alpízar, Róger Madrigal, and Lizette Delgado. Today it has grown into a well-known think-tank and I cannot express enough the great honor I feel being part of their team. Thank you Francisco Alpízar, Róger Madrigal, Juan Roablino, Carlos Muñoz, Catalina Sandoval, Rebecca Osakwe, Milagro Saborío-Rodríguez, Iréne Burgues, Adriana Chacón, Maria-Angelica Naranjo, and Laura Villalobos! A special thanks to Lizette Delgado, Andrea Castro, Alberto Vargas, Carlos, Laura, the enumerators Mille, Buenaventura, Jennifer and Layli, as well as the transportation unit at CATIE, for your tremendous help with the field work of this thesis. Further, this field work would not have been possible without the kind support from FONAFIFO, and the cooperation of landowners in Costa Rica. The financial support from the Tinker foundation made the data collection possible – thank you for believing in our project. I also gratefully acknowledge this incredible research environment made possible through the financial support from Sida (Swedish Agency for International Development and Cooperation) to the Environmental Economics Unit at the University of Gothenburg, via the Environment for Development Initiative (EfD).

Many thanks go to my graduate student colleagues and friends at the Department of Economics for interesting conversations and for making every moment more enjoyable. Special thanks to Lisa Andersson, Kristina Mohlin, Haileselassi Medhin, Hailemariam Teklewold, Simon Wagura, Jorge Bonilla, Michele Valsecchi, Xiaojun Yang, Qian Weng, and Claudine Uwera. Also great thanks to Yonas Alem, Eyerusalem Siba, and my colleagues and friends Laura and Maria-Angelica in Costa Rica who have not only supported me in my work but also helped me out whenever I needed a friend to call on.

(9)

Above all, I owe my deepest gratitude to my family – my mother for providing much-appreciated and respected pedagogical support and for being a loving mother, my father for his interest in my work and for supporting me by saying that anxiety is a normal part of earning a doctorate. Thank you my dear sister Sofie and my lovely niece Ellen for your love and support. My heartfelt thanks go to my love Fredrik for supporting me in the final stage of the thesis process – your love and happiness could take me over the highest mountain.

Finally, I remember when my grandmother aspired for me to become the prime minister of Sweden after I had expressed some concerns about the low earnings of retirees in the 1980s (I was 6 years old). And my grandfather eagerly asking: When will you become a professor? Even if I am not there yet, I know that you are very happy and proud of me now. Your high thoughts of me and your unconditional love have brought me to where I am today. This thesis is for you!

(10)

(11)

Summary of the thesis

Forests provide key ecosystem services such as clean water, timber, habitat for fisheries, carbon sequestration, pollination, and biodiversity. However, many of these services are being lost or degraded at a furious pace, brought about by human activity. For instance, deforestation and forest degradation are measured to account for around 12% of all CO2

emissions, making it the second largest anthropogenic source of carbon dioxide to the atmosphere after fossil fuel combustion (Werf et al. 2009). This has led communities, governments, and international organizations to increase their efforts to protect forests. Among such efforts, the use of monetary incentives to promote or reward private behavior that is associated with environmental objectives is becoming an increasingly popular policy instrument (Pattanayak et al. 2010, Ferraro 2011). For instance, payments for ecosystem services (PES) programs aim to increase the provision of ecosystem services by offering direct compensation to landowners for the opportunity costs of more environmentally friendly land management practices (e.g., low impact agriculture or conservation of natural ecosystems).

PES programs have been widely promoted as more cost-effective and institutionally less demanding than traditional conservation policies such as establishment of protected areas. Yet despite this, the few rigorous impact evaluations done so far show that the impact of PES programs has been modest (for a recent review see Pattanayak et al. 2010). This raises concerns that “easy fixes,” like PES, may not solve the planetary problems we are facing. Further, PES may suffer the fate of many interventions that stumble in reaching their objectives because people do not always behave as expected. Cardenas et al. (2000), for instance, experimentally show that introducing an incentive to reduce timber extraction from common forest land led to more forest extraction compared with a case with no incentive. This so-called crowding-out effect has been found in studies, both in psychology and economics, where external incentives sometimes lead to less pro-social behavior1 once the voluntary act is shifted to a market-based relationship (for a review of experimental as well as nonexperimental studies see Bowles 2008). At present, few attempts to understand the behavioral issues of forest conservation polices are undertaken.

This five-paper thesis attempts to contribute to the understanding of people’s behavioral responses to forest conservation policies. The first paper examines determinates of the impact

(12)

of payment for ecosystem services (PES) and the role of behavioral aspects. The second and third papers experimentally examine behavioral responses to incentives for voluntary contributions to forest conservation, where some stakeholders are excluded in favor of others. The fourth paper investigates the relationship between participation in PES programs and type of payment (i.e., cash or in-kind). The fifth and last paper examines the effect of introducing fixed entrance fees on voluntary donations to a protected area.

Paper 1: Incentives, Impacts, and Behavioural Issues in the Context of Payment for Ecosystem Services Programmes: Lessons for REDD+ (Published 10 April 2013 in Globalization and Development: Rethinking Interventions and Governance, A. Bigsten (Ed.), Routledge Press)

Payment for environmental services (PES) aims to increase the provision of public goods and internalize environmental externalities by offering direct compensation to landowners for the opportunity costs of more environmentally friendly land management practices (e.g. low impact agriculture or conservation of natural ecosystems). Being promoted as more cost-effective and institutionally simpler than traditional environmental conservation policies, mainly small-scale PES schemes have spread prolifically across developing countries in the last decade. Despite their popularity, there are few rigorous impact evaluations of existing PES programmes, and the ones that have been done have generally shown modest impacts. Here we use a conceptual framework of PES additionality, i.e. a programme’s ability to deliver outcomes that would not have occurred in its absence, to overview the main issues raised regarding the impacts of PES programme. We also show that PES impacts can be highly affected by information asymmetries and behavioural responses to the introduction and design of payment schemes. We draw upon these lessons to give policy advice to the design of REDD+ programmes.

Paper 2: Incentivizing versus Rewarding Good Behavior: Insights on the Use of Monetary Incentives

(13)

resulting in the exclusion of some stakeholders in favor of others. In this paper, we study the possibility of the stakeholders excluded from the monetary incentive reducing their pro-social behavior. We use a laboratory experiment to investigate this and hypothesize that alternative selection rules, i.e., who gets paid and why, affect the overall contributions to a public good differently. Our results show that incentivizing those who acted less pro-socially (i.e., contributed below a certain threshold) before the incentive was introduced resulted in increased contributions to the public good by this group. On the other hand, that very same selection rule excludes those who acted more pro-socially (i.e., contributed over a certain threshold) before the incentive was introduced, and this resulted in decreased average contributions by this group, decreasing the net effect on overall contributions. These results set up an efficiency-fairness tradeoff for designing selective conditional payments to promote pro-social behavior: Targeting those who require incentives to contribute may increase payment response beyond what would have happened in the absence of the incentive program, but it may also give rise to the unexpected consequence of negative spillovers.

Paper 3: Unintended Consequences of Targeting Forest Conservation Incentives: Behavioral Insights into Incentive Design

(14)

Paper 4: Payments in Cash or in Kind for Ecosystem Services: Stated Preferences of Costa Rican Landowners

This paper investigates landowners’ preferences for type of payment, cash or in kind, for the provision of ecosystem services. A choice experiment analysis focusing on the effect of different levels of cash and in-kind payments on participation in a payment for ecosystem services (PES) contract is provided. We use an educational in-kind payment in the form of days of practical training offered free of charge to the recipients. The results indicate a positive correlation between participation in a PES contract and the magnitude of the cash payment―higher cash payments increase the probability of participation—while participation seems uncorrelated with the magnitude of the in-kind payment. We also find that both in-kind and cash payments increase the likelihood of participation in shorter PES contracts (i.e., 5 years), while in-kind payments have no significant effect on participation in longer contracts (i.e., 15 years). Higher levels of cash payment seem to be what is needed to increase the likelihood of participation in longer contracts. In addition, we investigate heterogeneity in preferences for type of payment, which can help policymakers better target payment types to specific groups of landowners.

Paper 5: Do Entrance Fees Crowd Out Donations for Public Goods? Evidence from a Protected Area in Costa Rica

(15)

References

Bowles, S. 2008. Economic Experiments Undermine "The Moral Sentiments": Evidence from Policies Designed for Self-Interested Citizens May. Science 1605.

Cardenas, J. C., J. Strandlund, and C. Willis. 2000. Local Environmental Control and Institutional Crowding-Out. World Development 20:1719-1733.

Ferraro, P. J. 2011. The Future of Payments for Environmental Services. Conservation Biology 25:1134-1138.

Pattanayak, S. K., S. Wunder, and P. J. Ferraro. 2010. Show Me the Money: Do Payments Supply Environmental Services in Developing Countries? Review of Environmental Economics and Policy 4:254-274.

(16)

(17)

(18)

(19)

1

Published 10 April 2013 in Globalization and Development: Rethinking Interventions and Governance, A. Bigsten (Ed.), Routledge Press.

Incentives, Impacts, and Behavioural Issues in the Context of Payment

for Ecosystem Services Programmes: Lessons for REDD+

Anna Nordén, U. Martin Persson and Francisco Alpízar*

Payment for environmental services (PES) aims to increase the provision of public goods and internalize environmental externalities by offering direct compensation to landowners for the opportunity costs of more environmentally friendly land management practices (e.g. low impact agriculture or conservation of natural ecosystems). Being promoted as more cost-effective and institutionally simpler than traditional environmental conservation policies, mainly small-scale PES schemes have spread prolifically across developing countries in the last decade. Despite their popularity, there are few rigorous impact evaluations of existing PES programmes, and the ones that have been done have generally shown modest impacts. Here we use a conceptual framework of PES additionality, i.e. a programme’s ability to deliver outcomes that would not have occurred in its absence, to overview the main issues raised regarding the impacts of PES programme. We also show that PES impacts can be highly affected by information asymmetries and behavioural responses to the introduction and design of payment schemes. We draw upon these lessons to give policy advice to the design of REDD+ programmes.

____________________________________________________

* Anna Nordén, Department of Economics, University of Gothenburg, Sweden and EfD-Central America

at the Tropical Agriculture Research and Higher Education Center (CATIE), Costa Rica, (e-mail)

anna.norden@economics.gu.se; U. Martin Persson, Gothenburg Centre of Globalization and

Development, Department of Economics, University of Gothenburg, Box 640, 405 30 Göteborg, Sweden,

(email) martin.persson@economics.gu.se ; and Francisco Alpizar, EfD-Central America at the Tropical

Agriculture Research and Higher Education Center (CATIE), CATIE Headquarters 7170, Cartago,

Turrialba 30501, Costa Rica, (email) falpizar@catie.ac.cr .

The authors would like to thank Arne Bigsten and seminar participants in the workshop at Arken in Gothenburg for valuable comments and insights. Financing of the research presented in this chapter from the MISTRA programme BECC (Biodiversity and Ecosystem services in a Changing Climate), the Formas programme COMMONS, the Tinker Foundation and from SIDA (Swedish International

Development Cooperation Agency) via the Environment for Development Initiative

(20)

2

Introduction

The last decade has witnessed a rapid increase in the popularity—both in theory and practice—of payments for ecosystem services (PES) as an environmental policy tool in developing countries (Pattanayak et al. 2010, Ferraro 2011). An early review by Landell-Mills and Porras (2002) found approximately 200 incipient PES schemes in developing countries, and the numbers have only increased since then (Pattanayak et al. 2010). Although often small in scale, a few countries have established nationwide PES schemes: Costa Rica has its Pagos por Servicios Ambientales (PSA) programme, which since its inception in 1997, has made payments for forest conservation (primarily) on nearly half a million hectares of land; China has its Sloping Lands Conservation Programme (SLCP), which has thus far contracted 12 million hectares for reforestation in an attempt to stem soil erosion; and Mexico with its Pago de Servicios Ambientales Hidrológicos (PSAH) programme, which compensates beneficiary communities for preserving 600,000 hectares of forest (Pattanayak et al. 2010).

By directly compensating resource users for the opportunity costs of ecosystem service provision, PES has been touted as institutionally simpler and more cost effective than other, more indirect, conservation policies (Pattanayak et al. 2010, Ferraro 2011). Despite the claim that PES programmes are cost effective, few rigorous evaluations of the environmental and social impacts of PES programmes have been conducted. A recent review of the few credible assessments available concluded that programmes generally show poor performance in terms of delivering additional ecosystem services, and that ‘we do not yet fully understand either the conditions under which PES has positive environmental and socioeconomic impacts or its cost-effectiveness’ (Pattanayak et al. 2010:268).

(21)

3

forest carbon stock, with financing coming either from global carbon markets or from international funds.

Given its focus on performance-based payments, REDD+ is viewed as a multilevel PES scheme, where national and subnational PES programmes will be key tools for REDD+ implementation (Bond et al. 2009, Angelsen 2010, Pattanayak et al. 2010). If experience with PES programmes is to serve as a blueprint for REED+ implementation, it is important to understand the determinants of PES programme success and how outcomes are affected by information asymmetries and behavioural responses.

For the purposes of this book, it is important to recognise that most PES schemes implemented thus far have been local (e.g., county or municipality) or national initiatives, although, at times, these efforts have been supported by multilateral organisations. In the case of REDD+, funding for policy interventions comes from multilateral or bilateral initiatives; examples of the former include UN-REDD and the World Bank’s Forest Carbon Partnership Facility and Forest Investment Program, while Norway is the main funder of bilateral REDD+ activities (for an overview of existing REDD+ funding initatives see, for example, Westholm 2010). REDD+ is therefore understood here as a nationally implemented, but externally motivated and funded, intervention.

In this chapter, we describe a key issue raised in the literature regarding PES impact and efficiency: the extent to which PES programmes are capable of delivering desired outcomes beyond what would have occurred in their absence (referred to as additionality). This is a key issue for REDD+, because, unlike many past payments schemes, countries will only be eligible for payments if they reduce greenhouse-gas emissions from land-use change below an established baseline.1 The aim of this chapter is to offer guidance to policymakers regarding the circumstances under which PES is an appropriate policy choice and how PES programmes can be designed to maximise impact and minimise unwanted spillover effects.

(22)

4

how agents respond to economic incentives in general, including specific information resulting from economic experiments that examine behavioural response in relation to PES.

Accounting for behavioural responses is important because an economic agent’s reaction to a PES intervention is the result of a complex decision-making process that is only affected by the payment itself to a limited extent. For instance, monetary incentives could crowd in or out intrinsic motivations for protecting the environment. Additionally, participation in a PES programme might be due to peer pressure, learning, or simply inertia, e.g. neighbour and signalling effects, all of which are combined with the actual reaction to the payment.

Moreover, using both the model and insights from the behavioural economics literature, we discuss the ability to overcome the information asymmetries that limit programme efficiency through programme design, primarily by improving payment targeting.

Understanding the determinants of PES additionality: introducing a conceptual framework

Payment for ecosystem services is a policy that aims to increase the provision of ecosystem services and protect the natural-resource base by paying landowners for good agricultural practices or complete conservation of natural vegetation on their lands. If carefully designed and implemented, payments should provide sufficient incentives to adopt land-use management practices that reduce downstream pollution (e.g. reduced use of pesticides), avoid deforestation and increase carbon sequestration.

(23)

5

Following Persson and Alpízar (2011), given the voluntary nature of PES programmes, potential participants can be divided into four categories (see Figure 8.1): A — those who apply for payments, but will meet the programme conditions with or without them;

B — those who apply for payments and will not fulfil the conditions without payments; C — those who do not apply for payments but will meet the conditions regardless; and D — those who do not apply and will not meet the conditions.

Will meet PES conditions in absence of payment?

Yes No Appli es f or payment ? Yes

A:

UC(MC, NC) > UNC(MNC, NNC)

B:

UC(MC, NC) < UNC(MNC, NNC) UP(MP, NP) > 0 UP(MP, NP) > UNC - UC No

_C:

UC(MC, NC) > UNC(MNC, NNC)

_D:

UC(MC, NC) < UNC(MNC, NNC) UP(MP, NP) < 0 UP(MP, NP) < UNC - UC

Figure 8.1: Conceptual categorisation of potential PES payees based on their counterfactual compliance and application decisions. Agents will fall into categories A, B, C and D, depending on the utility (U) they would derive from meeting the programme conditions (C), the utility from non-compliance (NC), and the utility from PES participation (P), each of which is shaped by both monetary (M) and non-monetary (N) factors.

(24)

6

extent to which one is able to identify agents who will not meet conditions in absence of payments and use this information to target payments to them.

In the case where payments are not targeted based on the predicted risk of non-compliance, for example, deforestation risk (which is the case in the majority of existing PES schemes), additionality is strictly determined by (1) and (2); thus, (1) the higher the counterfactual compliance level, the lower the expected additionality, and (2) the more that selection bias causes compliers to self-select into the programme, the lower the additionality. Figure 8.2 illustrates these insights with results from a stylised multi-agent model of PES, simply being a numerical implementation of the conceptual framework presented in Figure 8.1.3

Figure 8.2 clearly shows that PES additionality and, consequently, cost-effectiveness will be seriously constrained in cases where counterfactual compliance with programme conditions is high. Another way to frame this result is in terms of information asymmetries (Pattanayak et al. 2010); to effectively implement PES, the regulator needs to target payments to those who would not meet the programme conditions in the absence of payments. However, this information is not available to the regulator but only to the PES applicants. The smaller the share of the eligible population that will not comply in the absence of payments, the larger this problem of hidden information is. Consequently, one cannot make general claims about the cost-efficiency of PES programmes; cost efficiency will be highly affected by the context in which a PES programme is implemented.

(25)

7

Figure 8.2: Results from a stylised multi-agent model of PES additionality, showing how additionality depends on counterfactual compliance (the share of potential payees that would have met programme conditions in the absence of payments), negative selection bias (those who already meet conditions self-select into the programmes at higher rates than others), and targeting based on imperfect predictors of counterfactual compliance. In the absence of selection bias and targeting, additionality roughly equals counterfactual non-compliance; negative selection bias reduces programme impacts, whereas targeting increases it. Good, medium, and poor targeting predictors correspond to cases where 10, 30, and 60%, respectively, of the variance in the variable that determines agents’ compliance decisions is determined by factors observable to the regulator. Low, moderate, and high selection bias correspond to cases where the correlation between the variables that determine agents’ decisions regarding compliance and participation are set to -0.3, -0.6, and -0.9, respectively. See (Persson and Alpízar 2011) for a full description of the model. The results from impact evaluations of the national PES programmes in Costa Rica (Pfaff et al. 2008, Robalino et al. 2008) and Mexico (Muñoz-Piña 2010, Alix-Garcia et al. 2012) are also shown.

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0,0% 20,0% 40,0% 60,0% 80,0% 100,0% A ddi ti ona lit y Counterfactual compliance

Mexico (PSAH) Costa Rica (PSA)

Target - good predictor Random - low selection bias

Target - medium predictor Random - moderate selection bias

Target - poor predictor Random - high selection bias

(26)

8

Selection bias can occur for two reasons, the first being that agents who will comply with the programme conditions in the absence of payments have a lower threshold for applying for payments than those who will not comply, as, in the latter case, the payments must cover the opportunity costs of compliance (see Figure 8.1). The second reason for selection bias may be that the factors that affect whether agents meet the programme conditions also affect the decision of whether to apply for payments.

In the Costa Rican case, selection bias seems to be primarily explained by the first mechanism: Ortiz et al. (2003) find that PES is only profitable on marginal lands with zero opportunity cost of conservation. Through interviews with both PSA participants and nonparticipants, Arriagada et al. (2009:355) find that the most common reason for enrolling land in PSA is a ‘lack of more profitable land use alternatives due to land characteristics’, and the second most common reason for not enrolling land — after lack of information — is that the payments are too low.4 In the Mexican case, however, poor programme design seems to be the chief reason for negative selection bias: land in the two quintiles with the highest estimated deforestation risk constitutes only 18 per cent of forest land eligible for conservation payments (Muñoz-Piña et al. 2008).

(27)

9

Beyond direct monetary incentives: spillovers from the introduction and design of ecosystem payments

There is an extensive literature on the factors affecting landowner attitudes towards environmental conservation and decisions regarding participation in PES-like systems (see reviews by, e.g., Siebert et al. 2006, Knowler and Bradshaw 2007). However, the latter studies almost exclusively focus on farmers’ participation in agro-environmental schemes (AES)5 in the EU, USA, and Australia; exceptions are studies from Costa Rica (Zbinden and Lee 2005, Wünscher 2008, Arriagada et al. 2009), Mexico (Kosoy et al. 2008), and China (Mullan and Kontoleon 2009). In the following, we will briefly review the findings from this literature, framing it with the conceptual model of PES additionality introduced in the previous section.

Under what circumstances will landowners fall into the different categories (A, B, C, and D) identified in the conceptual model? Starting with the decision of whether to comply with the programme conditions in the absence of payments, if we assume that agents maximise their expected utility6, we would expect landowners to comply if the (expected) utility they derive from doing so is larger than the (expected) utility from not complying. Studies analysing the factors affecting farm and forestry management decisions and PES participation have shown that these are not exclusively shaped by monetary incentives (e.g., Lynne et al. 1988, Siebert et al. 2006, Knowler and Bradshaw 2007). It is therefore, useful to make the distinction between monetary and non-monetary factors affecting the utility of compliance and PES participation.

Here, we will express the utility of complying with programme conditions or not as UC(MC, NC) and UNC(MNC, NNC), respectively, where MC is the land-use revenue under

compliance (e.g. incomes from non-extractive forest uses), NC is the non-monetary

(intrinsic) value of compliance (e.g. concerns for environmental issues, culture/spirituality reasons, signalling as environmental stewards), MNC is the land-use

revenue when not complying (e.g. incomes from agriculture or cattle ranching), and NNC

is the non-monetary (intrinsic) value of non-compliance (e.g. professional pride or traditional values of being a farmer or rancher). An agent will comply with PES conditions in the absence of payments only if UC > UNC (see Figure 8.1).

(28)

10

and non-monetary (NP) factors, UP(MP, NP). An agent will apply for, or accept,

payments if the utility from doing so is positive and offsets the opportunity cost of participation, i.e. if UP > max[0, UNC - UC]. The monetary incentive for participation

will depend on the payment level, the costs of participation (e.g. protection costs such as firebreaks in the case of forest conservation), and transaction costs (e.g. the costs of information gathering and contract establishment). The non-monetary benefit of PES participation may be either positive or negative as a result of, for example, pride of being recognised as an environmental steward or mistrust towards the regulator.

Consequently, a land-user’s decision of whether to apply for PES payments and conserve a patch of forest or to convert it to other uses will be shaped by both the economic returns to each option and the intrinsic well-being the landowner derives from having his or her land in either use or from participating in the scheme. As noted above, the empirical evidence suggests that monetary concerns are the primary motives for conservation and PES participation decisions, with the most common reason for enrolling land in PES in both developed- and developing-country settings being that the opportunity cost of doing so is low (a lack of profitable alternatives, fits with existing management plans), whereas a common reason for not enrolling land is that the payments are too low (Wilson and Hart 2000, Arriagada et al. 2009).

However, there is ample evidence that non-monetary factors also play a significant role in shaping these decisions. Studies across developed and developing countries on the adoption of environmentally friendly farming technologies in the absence of PES-like policies find that non-monetary factors — e.g. environmental awareness and concerns — increase conservation efforts (Lynne et al. 1988, Knowler and Bradshaw 2007). Similarly, Chouinard et al. (2008) find evidence of farmers in the US being willing to forgo some profit in the service of conservation objectives.

(29)

11

al. 2011). Wüncher et al. (2011) find that factors related to risk perception (risk aversion, age, trust in the regulator) affect farmers’ attitudes towards enrolling land in the Costa Rican PES programme.

Unfortunately, the interaction between the monetary and non-monetary attributes of a conservation policy is seldom studied. A common assumption is that utility is separable in the monetary and non-monetary arguments (Bowles 2008), such that a change in the level of one (e.g. the introduction of an economic incentive through a PES) does not affect the other. However, a growing body of both theoretical and experimental evidence shows that separability does not always hold, which can change both people's perceptions of a task as well as the outcomes of policy interventions, in sometimes unexpected ways (Frey and Jegen 2001, Bowles and Polania Reyes 2009). As a consequence, the introduction of conservation payments will not unequivocally raise the provision of ecosystem services.

In the next subsections, we discuss different ways in which indirect monetary and non-monetary motivations may affect the impact of introducing payments for ecosystem services. We begin by making a distinction between price spillovers and behavioural spillovers. Price spillovers are the side effects of introducing payment schemes that result from the change in relative prices that is caused by setting a portion of the available land aside for conservation. Behavioural spillovers are side effects from the PES scheme that affect land-use decisions by changing the personal motives for protecting a natural ecosystem, even in the absence of price changes.

Price spillovers: leakage from conservation efforts

(30)

12

Some recent empirical evidence for the Mexican PSAH programme found evidence of an approximately 4 per cent reduction in programme efficiency due to leakage, as a result of increased deforestation both on property belonging to the programme recipients and within markets with high levels of programme participation (Alex-Garcia et al. 2012). Wu et al. (2001) use an analytical framework and show that ignoring the effects of conservation strategies on prices can reduce the environmental gains.

Conceptually, as the prices of agriculture commodities and forest-extracting activities increase, the monetary revenue of non-compliance activities (MNC) increases,

potentially having a negative effect on the programme’s outcomes through leakage in the following ways: (1) turning As into Bs, threatening the outcomes of the PES by making compliance conditional on payment; (2) turning As or Cs into Ds, increasing non-compliance.

Although spillovers commonly are assumed to arise owing to changes in the relative value of land, it is important to add that non-pecuniary incentives might also be behind the observed leakage. For example, signalling could also play a role, where, by committing to protect one hectare of land, the landowner might feel entitled to clear another hectare somewhere else, without losing his/her green image. Alternatively, owners of large forests might be willing to commit to conservation on part of their land, in exchange for lenient tax treatment and environmental policy enforcement on the rest of their land.

Moreover, Robalino and Pfaff (2012) show that neighbours’ land-use decisions are significantly affected by each other. Using highly explicit spatial deforestation information, they show that neighbouring deforestation significantly increases the probability of deforestation. Such positive spatial interaction is good news, as policies promoting conservation in one area could potentially increase conservation in neighbouring areas. However, as mentioned above, the ways in which agents react to a policy will depend on how the policy is perceived.

Effects on the intrinsic valuation of compliance: motivation crowding and preference formation

(31)

13

the forest, presenting the villagers with the social dilemma of maximising their own payoffs and those of the community, as higher levels of wood extraction were posited to increase soil erosion and damage local water quality. As other public-goods games have shown, the villagers exhibited some other-regarding preferences by making decisions that were neither privately nor socially optimal, but somewhere in between. In a second round of the experiment, a regulation (an imperfectly enforced time quota on firewood gathering, coupled with a fine for noncompliance) was introduced. Although this initially reduced resource use, after a while, as participants realised that some participants where violating the quota and not getting caught, the time spent in the forest rebounded to its earlier value. The monetary incentive introduced by the regulation, therefore, did not succeed in increasing the provision of the public good, but simply replaced — or crowded out — the subjects’ intrinsic motivations for limiting resource use.

This experiment illustrates some general insights from the behavioural economics literature. First, agents tend to strike a balance between monetary and non-monetary interests. Second, external interventions can have unanticipated effects on the latter. With respect to PES, when an monetary incentive in the form of PES payments is introduced, voluntary compliance with the programme conditions becomes more of a market-like interaction, possibly affecting the non-monetary (intrinsic) motivation to comply with the programme conditions (cf. Heyman and Ariely 2004). The effects this may have on programme impacts can be both positive and negative, depending on how the monetary and social preferences interact.

The payment and the intrinsic motivation could be either (1) complements, implying that the introduction of a monetary incentive for conservation will increase the intrinsic motivation for the same, commonly called crowding in, or (2) substitutes, where the intrinsic values, as in the experiment above, are crowded out by a monetary incentive (Frey and Jegen 2001, Bowles and Polania Reyes 2009). In our conceptual model, motivation crowding would imply that, as we introduce (or change the level of) a payment, and therefore the level of UP, we will also affect the value of NC and therefore

UC.

Crowding out would lower UC, potentially having a negative effect on programme

(32)

14

the resulting Bs cannot be paid and by threatening the outcome of the PES in the long run by making compliance conditional on payments; and (2) by turning As, Bs, or Cs into Ds, which in all three cases increases non-compliance.

Conversely, the crowding in of social and environmental values, raising UC, could

potentially have a positive effect on programme compliance by (1) turning Ds into As, Bs, or Cs, and (2) by turning Bs into As. The latter will tend to increase compliance when not all PES applicants can be paid and would have a positive long-term effect on preferences for environmental conservation.

Moreover, as individuals are also motivated partly by how they are perceived by others, forest conservation may well be partly motivated by preferences for being perceived as a pro-environmental person. This so-called image or signalling motivation captures the rule of opinion in utility, i.e. the desire to be liked and respected by one’s peers. Agents therefore, attempt to signal characteristics that are defined as ‘good’, based on social norms and values, in the search for social approval (for economic models incorporating social approval, see, for example, Bénabou and Tirole 2006, Andreoni and Bernheim 2009, Ellingsen and Johannesson 2008).

Studies have found that such image motivation could be crowded out when an incentive is introduced. In an experimental study, Ariely et al. (2009) found a negative relationship between payments and image motivation, where payments had a tendency to crowd out the motivation to signal socially preferable behaviour. A possible explanation for this result is that the signalling effect becomes unclear, as some individuals are behaving in accordance with social norms simply owing to the incentive. Introducing a PES payment conditional on forest conservation may then crowd out the green-image motivation, as it becomes impossible to distinguish between those conserving the forest because of the payment and those that would have done so in any case.

(33)

15

(via monetary donations to an environmental charity) when the conservation action is removed, suggesting motivational crowding out.

In a stated-preference survey among Danish forest owners, Boon et al. (2010) find a small crowding-out effect, with 2 per cent of the sample of forest owners being willing to set aside less of the forest for conservation under a payment scheme than without it (although this minor negative effect is overwhelmed by the large positive effect on the willingness to conserve forest land among the rest of the sample). There are a few studies that find that participation in PES in the EU induces some changes in preferences, suggesting crowding in, but overall attitudes towards environmental conservation seem to be unaffected by participation in PES schemes, leading Burton et al. (2008: 18) to conclude that ‘the schemes act as a facilitator for the expression of existing attitudes rather than agents of attitude change’.

In a comparative study of three PES schemes in Latin America, Kosoy et al. (2008) find that payments often do not cover the farmer’s opportunity cost of participation, with participation partly being explained by participants feeling that the PES supports them as environmental stewards. Although this does not constitute proof of crowding in — merely that considerations other than purely pecuniary ones play a role — it does indicate that there is no strong effect in the opposite direction.

A small but positive crowding effect from a voluntary, positive economic incentive, such as a PES, is consistent with the psychological notion that crowding in occurs when agents feel that external interventions are supportive, whereas crowding out occurs when agents feel that interventions are designed to control their behaviour (Frey and Jegen 2001) (cf. results from Cardenas et al. 2000 discussed above). However, as will be discussed below, a positive effect may be contingent on participants perceiving the payment scheme as fair (Bowles and Polania Reyes 2009), something that may be affected by measures to increase programme impacts through, for example, increased targeting.

(34)

16

concern here may, therefore, be that there is little evidence to date supporting the notion that these programmes affect the long-term preference formation of participants, which implies that, to maintain the environmental gains from PES, payments must continue indefinitely. Consequently, if the incentives for deforestation increase in the future — owing to increased demand for land for food, feed, and biofuel — PES may have done little to buttress the support for forest conservation, and whatever gains that have been made may be lost (Persson 2012).

Fairness versus efficiency: the potential for, and limits to, targeting payments to increase additionality

Past experience clearly shows that failing to target risk has constrained the impacts on forest conservation efforts, both in the national PES programmes in Costa Rica and Mexico and regarding protected areas, the locations of which have been found to be biased towards areas unlikely to face land-conversion pressures (Andam et al. 2008, Joppa and Pfaff 2009). However, the results also show that programme design does matter: the Costa Rican PSA system achieved greater additionality in the 2000-2005 period thanks to new selection rules that reduced bias towards low-risk landowners (Robalino et al. 2008).

Several studies acknowledge the need for improved PES targeting to counteract adverse selection and increase additionality (Pattanayak et al. 2010, Ferraro 2011). However, although targeting has the theoretical potential to substantially increase the efficiency of PES programmes, there are a number of obstacles that may hinder its success in practice. The foremost is, of course, the reason that PES may perform poorly in the first place: the issue of asymmetric information. As landowners know more about their ecosystem service’s vulnerability and opportunity costs than do PES regulators, they are able to extract informational rents from PES buyers in the form of payments for non-vulnerable ecosystem services or payments well above the full costs of protection (including direct costs, transaction costs, and opportunity costs).

(35)

17

implement targeting strategies, which, in practice, can be quite expensive (Engel et al. 2008, Ferraro 2008).

A simpler, and less expensive, way to target payments that does not require information about non-compliance risk for individual applicants is by restricting a PES programme to an area with a higher incidence of baseline non-compliance. Both in the Amazon and in South East Asia, annual deforestation rates may exceed 5 per cent in some areas (Mena et al. 2006, Miettinen et al. 2011), implying non-compliance rates — and, consequently, potential PES additionality — that are an order of magnitude higher than those prevailing at national levels.

However, the gains from both geographic and individual targeting may be offset through increased market-based leakage, with the deforestation pressure simply shifting to areas or individuals not targeted by the programme. Moreover, targeting almost inevitably causes PES policies to treat certain groups of potential participants differently. It follows that selection rules, differentiated payments and other targeting strategies can introduce questions about the equity and fairness of PES programmes and, consequently, the risk of what one can call behavioural spillovers.

Economic theory regarding fairness suggests that people in general are inequity averse; i.e. agents experience disutility, not only from being worse off than others, but also from being better off (Fehr and Schmidt 1999, Bolton and Ockenfels 2000). The evidence reveals, not simply that inequity aversion is distaste for inequity, but that agents are willing to incur some private cost to avoid it.

In the context of PES, a narrow focus on efficiency will most likely suggest targeting payments to those acting in their own self-interest to shift their behaviour. This would imply paying agents that are, in the words of Wunder (2007: 53), ‘if not outright environmentally nasty, then at least at the edge of becoming so’. Although potentially efficient, when we consider only programme beneficiaries, this sort of selection rule might lead those not being paid, owing to their perceived low risk of non-compliance to feel unfairly treated. A possible result of this may be that they retaliate by deforesting their land, justified by the feeling that they were not rewarded but punished for their previous environmental stewardship (Lindhjem et al. 2011).

(36)

18

over 400 students in total. The basic structure of a dictator game is that the dictator receives an endowment to divide between himself and a receiver. Their results concern a three-period game, in which all subjects initially play a dictator game where the receiver is a green public fund used to protect forests in Costa Rica (i.e. before the policy is created). When the second round is played, some players qualify for conditional payments, i.e. a reduced cost for contributions in the third period. One rule for who qualifies is a lottery. Another is that the people who contributed more (specifically above a threshold) in the second period received payments, and a third is that the people who contributed less (specifically below a threshold) within the second period received payments. The authors find that those not selected for payment reduce their contributions significantly depending on the selection rule. In particular, they find the greatest behaviour spillovers when payments go to the non-contributors, a result that is strengthened further by a natural field experiment performed by Alpízar et al. (2012b). This presents an efficiency dilemma for PES: targeting those who require incentives to contribute might lead to the desired response, but may also produce undesirable behavioural responses from those not selected for payment.

Returning to the conceptual model in Figure 8.1, targeting only those that comply with the programme’s conditions (Bs) may generate behavioural responses that could decrease the utility of compliance (Uc), as the non-monetary motivations (NC) for

complying for those that are not selected for payment might decrease. This would have a potentially negative effect on the impact of PES by tuning those who comply with the programme without payment, and do not receive a payment in this case (As and Cs), into landowners who would only comply if paid (Bs or Ds), potentially producing a negative effect if Bs cannot be paid, or in the long run if the payments will not continue indefinitely.

(37)

19

for their low contributions (for efficiency reasons) contributed more while those that were excluded from payment for their relatively high contributions contributed less.

The results from these experiments confirm the notion that, not only is there a trade-off between efficiency and fairness considerations in PES programmes, but that excessive emphasis on efficiency may cause negative spillover effects, undermining the very aim of the programme. In other words, behavioural responses and preferences for equity and fairness may very well place bounds on the additionality that can be achieved in PES programmes.

Policy discussion and conclusions: what is the proper role for PES in national REDD+ implementation?

In this final section we ask the following question: in light of the evidence presented above, what is the role for PES in implementing REDD+? The first important conclusion emerging from the conceptual framework is that the meagre performance of the national PES programmes in Costa Rica and Mexico is not primarily a result of poor policy design (although this has also diminished policy impact); rather, the fact that the level of additionality is on the order of 1 per cent or less simply reflects that non-compliance with the programme conditions (i.e. deforestation rates) are in the same order of magnitude.

The mirror argument of this, however, is that PES is potentially a cost-effective policy for inducing increased reforestation (which is a component of the plus in REDD+), given that baseline reforestation rates in many tropical countries are low. In fact, this seems to be the experience emerging from the Costa Rican PSA programme (Daniels et al. 2010), although a much smaller share of payments have been gone to reforestation contracts than forest conservation.

(38)

20

will be at work if payments are targeted to individual applicants based on non-compliance risk. Note also that, even in the absence of these offsetting effects, deforestation risks are inherently difficult to predict (especially if one can only rely on non-manipulative predictors to reduce the risk of moral hazard), and, consequently, the absolute gains from targeting may still be limited (see Figure 8.2).

These results have significant implications for the role that PES can play in implementing REDD+ in countries with tropical forests, given that the essence of REDD+ is performance-based payments. That is, only if a country can show that it has reduced emissions from deforestation and degradation below a given level will it be eligible for receiving REDD+ funds or selling REDD credits. This implies that, if a country selects a nationwide PES scheme as its main REDD policy and that scheme exhibits an additionality of 1 per cent, the payment level either has to be set at 1/100 of the international carbon price (which will most likely provide little incentive for forest conservation), or the country has to provide co-funding for payments to the 99 per cent of landowners who will be paid for doing what they would have done in the absence of the intervention (something that few developing countries’ budgets are likely to accommodate).

This does not mean that there is no place for PES in implementing REDD+. It does, however, imply that a nationwide PES programme most likely will not be the main instrument actually realising reduced deforestation and forest degradation, but that this requires a broader set of policies.

(39)

21

There are several benefits to implementing PES as a part of a larger set of policies aimed at forest conservation. If the main aim of PES is no longer cost efficiency, broader policy aims, such as poverty alleviation and fairness, can more easily be accommodated, as they no longer necessarily compromise the overall efficiency of the intervention. Other policies and measures at the national level, addressing the causes of deforestation and providing alternative livelihood options, may also reduce the risk of leakage that might otherwise occur in a standalone PES programme (Sunderlin and Sills 2012).

However, if the role of, and motivations for, PES change from cost-effective conservation policy to benefit sharing and increasing the legitimacy of other conservation policies, policymakers need to contemplate whether a PES scheme is the best option for meeting these objectives and be aware of the broader effects that the introduction of a payment scheme may have. Especially as this chapter has highlighted that the introduction of PES may have unintended effects on landowners’ intrinsic motivations for forest conservation and, therefore, undermine, rather than build, long-term support for forest conservation. This effect would be unfortunate for two reasons. First, REDD+ is only intended to be a temporary solution; in the longer term, international REDD+ financing will cease, and when this occurs, it is important that the institutions and policies implemented as part of REDD+ can be sustained. The question is whether tropical countries will be willing, and able, to continue large-scale PES schemes when external financing vanishes (this has been, and continues to be, a concern in Costa Rica where the PSA programme was initially funded, to a large extent, by a World Bank grant).

(40)

22

Notes

1_{However, establishing a baseline to ensure that estimated emission reductions are truly}

additional is far from straightforward (Persson and Azar 2007).

2_{An exception is the Chinese SLCP, where some involuntary enrolment has been}

reported (Bennett 2008). Sommerville et al. (2009: 2) argue that, although participation in a PES scheme is voluntary, ‘service providers do not necessarily have the choice whether or not to provide the service, such as in cases where land-use change is illegal.” However, such restrictions are seldom (if ever) perfectly enforced, and landowners may choose to deforest, even if such an action is illegal. It is estimated that roughly 85 per cent of all tropical deforestation occurs illegally.

3_{The model generates a random sample of 10,000 agents, each representing a potential}

PES recipient and characterised by the following: (1) the loss in utility from complying with PES conditions (UNC – UC in Figure 8.1); and (2) the utility derived from PES

participation (UP in Figure 8.1). If the former is positive, agent i will not meet the

conditions in the absence of payments. Similarly, agent i will only apply for payments if the utility derived from doing so is positive and covers the associated opportunity cost (see Figure 8.1). Both agent characteristics are assumed to be normally distributed, with expected means and variances chosen such that a given level of counterfactual compliance with the programme and share of agents applying for payments is achieved. To model selection bias, the two characteristics are set to be correlated, with a correlation coefficient s in the interval [-1,1], such that, if s < 0, there is negative selection bias, and, if s > 0, there is positive selection bias. The results presented are averaged over 1,000 runs. See Persson and Alpízar (2011) for a full description of the model.

4_{Similarly, in a developed-country setting, Wilson and Hart (2000) found that ranking}

(41)

23 5

The difference between AES and PES is largely semantic, and we will therefore refer to both types of schemes as PES here.

6_{Note that this does not presuppose that these perceptions are rational or that the}

resulting decisions are privately or societally optimal.

Bibliography

Alix-Garcia, J., E. N. Shapiro, and K. R. E. Sims. 2012. Forest conservation and slippage: Evidence from Mexico's national payments for ecosystem services program. Land

economics 88(4): 613-638.

Alpízar, F., A. Nordén, A. Pfaff and J. Robalino. 2012a. Incentivizing versus rewarding good behavior. Insights on the use of monetary incentives. Working paper.

Alpízar, F., A. Nordén, A. Pfaff, and J. Robalino. 2012b. Unintended consequenses of targeting forest conservation incentives: Behavioral insights into incentive design. Working paper.

Andam, K. S., P. J. Ferraro, A. Pfaff, G. A. Sanchez-Azofeifa, and J. A. Robalino. 2008. Measuring the effectiveness of protected area networks in reducing deforestation.

Proceedings of the National Academy of Sciences of the United States of America

105:16089-16094.

Anderson, C. L. 2006. Understanding the supply response of local public goods to environmental service payments. Working Paper 2006-03. University of Washington, Seattle.

Andreoni, J. and B. D. Bernheim. 2009. Social Image and the 50-50 Norm: A Theoretical and Experimental Analysis of Audience Effects. Econometrica 77:1607-1636. Angelsen, A. 2010. Policies for reduced deforestation and their impact on agricultural

production. Proceedings of the National Academy of Sciences 107:19639-19644. Angelsen, A., M. Brockhaus, M. Kanninen, E. Sills, W. D. Sunderlin, and S.

Wertz-Kanounnikoff. 2009. Realising REDD+: national strategy and policy options. Center for International Forestry Research (CIFOR), Bogor, Indonesia.

Angelsen, A., M. Brockhaus, W. D. Sunderlin, and L. Verchot. 2012. Analysing REDD+: Challenges and choices. Center for International Forestry Research (CIFOR), Bogor.

Ariely, D., A. Bracha, and S. Meier. 2009. Doing Good or Doing Well? Image Motivation and Monetary Incentives in Behaving Prosocially. American Economic Review

99:544-555.

Arriagada, R. A., E. O. Sills, S. K. Pattanayak, and P. J. Ferraro. 2009. Combining Qualitative and Quantitative Methods to Evaluate Participation in Costa Rica's Program of Payments for Environmental Services. Journal of Sustainable Forestry

28:343 - 367.

Bénabou, R. and J. Tirole. 2006. Incentives and Prosocial Behavior. American Economic

Review 96:1652-1678.

Bennett, M. T. 2008. China's sloping land conversion program: Institutional innovation or business as usual? Ecological Economics 65:699-711.

Bolton, G. E. and A. Ockenfels. 2000. ERC: A Theory of Equity, Reciprocity, and Competition. American Economic Review 90:166-193.

(42)

24

Boon, T. E., S. W. Broch, and H. Meilby. 2010. How financial compensation changes forest owners' willingness to set aside productive forest areas for nature conservation in Denmark. Scandinavian Journal of Forest Research 25:564-573.

Bowles, S. 2008. Policies designed for for Self-Interested Citizens may undermine "The Moral Sentiments": evidence from economic experiments. Science 320:1605-9. Bowles, S. and S. Polania Reyes. 2009. Economic Incentives and Social Preferences: A

Preference-Based Lucas Critique of Public Policy. CESifo Working Paper Series No. 2734

Burton, R. J. F., C. Kuczera, and G. Schwarz. 2008. Exploring Farmers' Cultural Resistance to Voluntary Agri-environmental Schemes. Sociologia Ruralis 48:16-37.

Cardenas, J. C. 2004. Norms from outside and from inside: an experimental analysis on the governance of local ecosystems. Forest Policy and Economics 6:229-241.

Cardenas, J. C., J. Stranlund, and C. Willis. 2000. Local Environmental Control and Institutional Crowding-Out. World Development 28:1719-1733.

Chouinard, H. H., T. Paterson, P. R. Wandschneider, and A. M. Ohler. 2008. Will Farmers Trade Profits for Stewardship? Heterogeneous Motivations for Farm Practice Selection. Land Economics 84:66-82.

Daniels, A. E., K. Bagstad, V. Esposito, A. Moulaert, and C. M. Rodriguez. 2010. Understanding the impacts of Costa Rica's PES: Are we asking the right questions?

Ecological Economics 69:2116-2126.

Ellingsen, T. and M. Johannesson. 2008. Pride and Prejudice: The Human Side of Incentive Theory. American Economic Review 93:990-1008.

Engel, S., S. Pagiola, and S. Wunder. 2008. Designing payments for environmental services in theory and practice: An overview of the issues. Ecological Economics 65:6 6 3 – 6 7 4.

Fehr, E. and K. Schmidt. 1999. Theory of Fairness, Competition, and Cooperation.

Quarterly Journal of Economics 114:817-868.

Ferraro, P. J. 2001. Global Habitat Protection: Limitations of Development Interventions and a Role for Conservation Performance Payments. Conservation Biology 15:990-1000.

Ferraro, P. J. 2008. Asymmetric information and contract design for payments for environmental services. Ecological Economics 65:8 1 0 – 8 2 1.

Ferraro, P. J. 2011. The Future of Payments for Environmental Services. Conservation

Biology 25:1134-1138.

Frey, B. S. and R. Jegen. 2001. Motivation Crowding Theory. Journal of Economic Surveys

15:589-611.

Hartshorn, G., P. J. Ferraro, B. Spergel, and E. O. Sills. 2005. Evaluation of the World Bank – GEF Ecomarkets Project in Costa Rica. Report of 'Blue Ribbon Evaluation Panel'. Hertel, T. W. 2010. The Global Supply and Demand for Agricultural Land in 2050: A

Perfect Storm in the Making? GTAP Working Paper No.63.

Heyman, J. and D. Ariely. 2004. Effort for Payment: A Tale of Two Markets. Psychological

Science 15:787-793.

Joppa, L. N. and A. Pfaff. 2009. High and Far: Biases in the Location of Protected Areas.

PLoS ONE 4:e8273.

Kaimowitz, D. 2008. The Prospects for Reduced Emissions from Deforestation and Degradation (REDD) in Mesoamerica. International Forestry Review 10:485-495. Kits, G., W. Adamowicz, and P. Boxall. 2012. Do conservation auctions crowd out

voluntary environmentally friendly activities? Working paper.

(43)

25

Kosoy, N., E. Corbera, and K. Brown. 2008. Participation in payments for ecosystem services: Case studies from the Lacandon rainforest, Mexico. Geoforum 39:2073-2083.

Landell-Mills, N. and I. T. Porras. 2002. Silver bullet or fools’ gold? A global review of markets for forest environmental services and their impact on the poor. International Institute for Environment and Development (IIED), London, UK.

Lindhjem, H., I. Aronsen, K. G. Bråten, and A. Gleinsvik. 2011. Experiences with benefit sharing: Issues and options for REDD-plus. Econ Report no. R-2010-018, Econ Pöyry, Oslo.

Lynne, G. D., J. S. Shonkwiler, and L. R. Rola. 1988. Attitudes and Farmer Conservation Behavior. American Journal of Agricultural Economics 70:12-19.

Mena, C., R. Bilsborrow, and M. McClain. 2006. Socioeconomic Drivers of Deforestation in the Northern Ecuadorian Amazon. Environmental Management 37:802-815. Miettinen, J., C. Shi, and S. C. Liew. 2011. Deforestation rates in insular Southeast Asia

between 2000 and 2010. Global Change Biology 17:2261-2270.

Mullan, K. and A. Kontoleon. 2009. Participation in Payments for Ecosystem Services programmes in developing countries: The Chinese Sloping Land Conversion Programme. Environmental Economy and Policy reseach Discussion Paper Series University of Cambridge.

Muñoz-Piña, C. 2010. Measuring the REDD effect of the payment for hydrological environmental services program in Mexico (PSAH). Workshop on enhancing cost-effectiveness of payment for ecosystem services. Organization for Economic Co-operation and Development (OECD). Paris.

Muñoz-Piña, C., A. Guevara, J. M. Torres, and J. Braña. 2008. Paying for the hydrological services of Mexico's forests: Analysis, negotiations and results. Ecological

Economics 65:725-736.

Ortiz, E., L. F. M. Sage, and C. C. Borge. 2003. Impacto del Programa de Pago de Servicios Ambientales en Costa Rica como medio de reduccion de la pobrexa en los medios rurales. Unidad regional de asistencia tecnica ruta.

Pattanayak, S. K., S. Wunder, and P. J. Ferraro. 2010. Show Me the Money: Do Payments Supply Environmental Services in Developing Countries? Review of Environmental

Economics and Policy 4:254-274

Persson, U. M. 2012. Conserve or convert? Pan-tropical modeling of REDD–bioenergy competition. Biological Conservation 146:81-88.

Persson, U. M. and F. Alpízar. 2011. Conditional Cash Transfers and Payments for Environmental Services: A Conceptual Framework for Explaining and Judging Differences in Outcomes. Environment for Development Working Paper, EfD DP 11-06:32 pp.

Persson, U. M. and C. Azar. 2007. Tropical deforestation in a future international climate policy regime—lessons from the Brazilian Amazon. Mitigation and Adaptation

Strategies for Global Change 12:1277-1304.

Pfaff, A., J. A. Robalino, and G. A. Sanchez-Azofeifa. 2008. Payments for Environmental Services: Empirical Analysis for Costa Rica. Working Paper Series, SAN-0805, Duke University, Terry Sanford School of Public Policy, Durham, NC, USA. Robalino, J. and A. Pfaff. 2012. Contagious Development. Neighbor Interactions in

Deforestation. Journal of Development Economics 97:427-436.